trading data maths anki reading

This page will be dedicated to analysis and visualisation of data.

Import libraries

import pandas as pd
import plotly.graph_objects as go
import datetime
from binance.client import Client
import numpy as np
import websocket
import json
import rel
import seaborn
import matplotlib as plt

Read CSV files and turn them into dataframes

• Insert the file names for data to be analysed
• K-line data in the format provided by binance is accepted

csv1 = 'bitcoin.csv'
csv2 = 'ethereum.csv'

Initialise client for Binance API

• we can get historical data using this instead
• documentation for Binance API
• official docs for the API

api_key = 'api_key'
api_secret = 'no no no'

client = Client(api_key, api_secret)

Data to fetch from Binance API

symbol1 = 'BTCUSDT'
interval1 = '1m'
start1 = ''
end1 = ''

symbol2 = 'ETHUSDT'
interval2 = '1m'
start2 = ''
end2 = ''

symbol3 = 'LTCUSDT'
interval3 = '1m'
start3 = ''
end3 = ''

Get data from the API and function to turn into dataframe

• format is (symbol, interval, start, end, limit)
• Pandas for dataframes

klines1 = np.array(client.get_historical_klines(symbol1, interval1))
klines2 = np.array(client.get_historical_klines(symbol2, interval2))
klines3 = np.array(client.get_historical_klines(symbol3, interval3))

def binanceDataFrame(klines):
    df = pd.DataFrame(klines.reshape(-1,12),dtype=float, columns = ('time', 'open', 'high', 'low', 'close', 'volume', 'close_time', ' quote_volume', 'count', 'taker_buy_volume', 'taker_buy_quote_volume', 'ignore'))
    df['time'] = pd.to_datetime(df['time'], unit='ms')
    
    return df

pair1 = binanceDataFrame(klines1)
pair2 = binanceDataFrame(klines2)
pair3 = binanceDataFrame(klines3)

Plot data using plotly

• Plotly for plotting candlestick charts

def candlestickplot(df):
    fig = go.Figure(data=[go.Candlestick(x=df['time'],
                open=df['open'],
                high=df['high'],
                low=df['low'],
                close=df['close'])])
    
    return fig

fig1 = candlestickplot(pair1)
fig2 = candlestickplot(pair2)
fig3 = candlestickplot(pair3)

Run below cell to plot candle stick charts

fig1.update_layout(title=symbol1, xaxis_rangeslider_visible = False)
fig2.update_layout(title=symbol2, xaxis_rangeslider_visible = False)
fig3.update_layout(title=symbol3, xaxis_rangeslider_visible = False)

fig1.show()
fig2.show()
fig3.show()

Getting live data using websockets and the Binance API

• websockets docs
• binance websocket stream docs

socket = f'wss://stream.binance.com:9443/stream?streams={symbol1.lower()}@kline_{interval1}/{symbol2.lower()}@kline_{interval2}/{symbol3.lower()}@kline_{interval3}'

    
def on_open(ws):
    print('---------Opened----------')
def on_message(ws, message):
    jsonmessage = json.loads(message)
    data = jsonmessage['data']
    symbol = data['s']
    candle = data['k']
    time = datetime.datetime.fromtimestamp(candle['t']/1000.0)
    open = candle['o']
    close = candle['c']
    high = candle['h']
    low = candle['l']            
    closed = candle['x']
    if closed:
        print(f'{symbol}: opened at {open}, closed at {close} at {time}')
    
websocket.enableTrace(False)
ws = websocket.WebSocketApp(socket, on_message = on_message, on_open = on_open)
ws.run_forever()

Running statistical analysis

• Seaborn for data visualisation

closes = pd.concat([pair1['close'], pair2['close'], pair3['close']], axis = 1)
closes.columns = [symbol1, symbol2,symbol3]
correlation = closes.corr()
heatmap = seaborn.heatmap(correlation, cmap="crest", annot = True)